1887.] Contributions to the Metallurgij of Bismuth, 89
Felruary 10, 1887.
Professor STOKES, D.C.L., President, in the Chair.
The Presents received were laid on the table, and thanks ordered
The following Papers were read : —
I. " Contributions to the Metallurgy of Bismuth." By EDWARD
Matthey, F.S.A., F.C.S., Assoc. Koy. Sch. Mines. Commu-
nicated by John Percy, M.D., F.R.S., Pres. Iron and Steel
Inst. Received August 18, 1886.
§ 1. Bismuth: its Separation from Gold, and its Refining Action
upon sams during the Process of Separation. — In bringing the above
subject under notice, it is necessary to allude to some of the facts dis-
tinguishing this very interesting metal.
Bismuth, in some of its important characteristics and reactions,
resembles lead. And one of the chief points of resemblance between
these metals is their ready oxidation, and their absorption by bone-
ashes or wood-ashes whilst so oxydised. I refer of course, to the
process of cupellation.
This ancient and serviceable process, still employed universally
for the separation of gold and silver from lead, is equally applicable
to bismuth, if associated with these precious metals ; and, like lead,
bismuth may be readily employed as a vehicle or means of collecting
gold and silver from their ores in reduction processes ; but its com-
parative cost accounts for its non- employment in this respect.
Commercially speaking, bismuth differs from lead in its greater
value, lead being wori3h at present £1S to £14 per ton,* whilst bismuth
realises between £700 and £800 per ton ; this high value being due to
its greater rarity and to its limited and special uses.
As is well known, bismuth ores are frequently auriferous ; and
one of the points which it is my desire to bring under notice is the
effectual separation of the gold from bismuth by a rapid and effica-
Of course, nothing could be easier than to separate these two
metals by the ordinary process of cupellation. The gold, by these
* June, 1886.
90 Mr. E. Mattliey. [Feb. 10,
means, is at once rendered ayallable, but wifcli the drawback that not
only is there a very considerable loss of bismuth by volatilisation
during the cupellation, but the subsequent recovery of the metal,
which in the state of oxide has been absorbed by the cupel, is
rendered necessarj, involving a tedious and troublesome smelting
operation, the employment of expensive fluxes, and a further con-
siderable loss of metal.
Bearing in mind the close resemblance of bismuth to lead in its
behaviour in the cupellation process, I directed my attention to its
separation from gold by means of the addition of a small proportion
of zinc — a method known as the Parkes process, as employed for the
separation of silver from lead. And this I found successful, the
natural separation of these two metals during the process of cooling
proving to be similar in both cases.
The operation as carried out by me is as follows : —
The Bismuth holding the gold is melted at the ordinary tempera-
ture, about two per cent, of melted zinc is then added, and the
whole brought to a dull red heat. The alloy is then well stirred,
and the temperature gradually lowered. When at a black heat the
slight crust formed on the surface is skimmed off and the metal
again treated with a further quantity of zinc at the higher tempera-
ture. The whole of the gold will be found in these skimmings, and
the bismuth v/ill be thus freed from it.
The skimming^s, consisting of bismuth, gold and zinc, and zine
oxide, I now treat by a process which quickly renders the gold avail-
able, and at the same time has the effect of refining the gold from all
impurities excepting silver during the actual process of extraction.
This small proportion of bismuth litharge and its charge of gold
is fused in a clay crucible with a little borax, and allowed to cool
down in the crucible, or it is poured into a mould with the bismuth,
litharge, which being perfectly liquid, allows the metallic gold to
separate by its own gravity, and during its fusion absorbs any base
metals associated with it as oxides. The bismuth litharge^ in fact,
acts as a refining agent to the goldj which, when cold, is detached from
it. This bismuth slag is broken up, re-fused with a little metallic
bismuth, and is so freed from the last trace of gold which is collected
by the bismuth, and subsequently extracted. The bismuth litharge
so freed from gold is then reduced by fusion with carbon to its
The quantity of bismuth litharge holding the gold is exceedingly
small in proportion to the bulk of metal originally treated, as the
figures hereinafter given will show ; but, by this process the bismuth
is at once freed from its gold contents with little time, labour, or
I have continuously carried out this method of treatment with the
1887.] Contributions to the Metallurgy of Bismuth. 91
most satisfactory results. It will only be necessary to take the figures
of one operation as an illustration.
A quantity of 9483 lbs. of bismuth, holding about one per cent, of
impurity, and 12'5 ounces of gold per ton (equal to 53*5 ounces in
the bulk), was so treated, and of this nearly 9000 lbs. was imme-
diately rendered available for commercial use, the skimmingv*?, which
amounted to 668 lbs. (7*30 per cent, of the bulk), contai^iing the tvJiole
of the gold.
These skimmings I oxidised by means of nitric acid, thus obtain-
ing the greater proportion of the bismuth and what little copper there
was in solution, from which the bismuth was precipitated by the
ordinary method, care being taken to saturate the nitric acid by ex-
tracting the greater portion of the bismuth as nitrate, so as to leave
a portion of the bismuth as oxide with the gold in order to refine
it from the impurities existing as oxides when fused with it. This
residue, collected and dried, was, when dried, fused in clay crucibles,
with a small quantity of borax, yielding the full amount of gold shown
A.8 before stated, in these fusions the metallic gold separates from
the bismuth litharges, and descends to the bottom of the crucible by
its own gravity. The liquid and supernatant bismuth litharge floats
upon it and breaks away readily when cold, the gold so obtained being
associated only with silver, both metals being in fact refined hy the
action of the bismuth litharge.
§ 2. Separation of Bismuth from Leac?.— The difficulty surrounding
the treatment of bismuth associated with other metals by any rapid
or comprehensive process is well known to the metallurgical chemist.
I believe I am correct in stating that hitherto the only process
employed for the refining of bismuth on the Continent — notably in
Saxony, the chief continental source of this metal — has been that of
chlorination and subsequent precipitation, a process tedious in itself
and involving much plant and labour in comparison with the quanti-
ties of metal operated upon.
Hapidity of production with a minimum margin of loss, in order
to free the metal from its impurities and render it marketable as
quickly as possible, being a great desideratum, induced me to turn
my attention to its refining by dry processes. In carrying this out I
have found present most of the metals which are easily seized by and
become associated with the bismuth itself during the process of reduc-
tion from its ores,* such as antimony, arsenic, tellurium, lead, copper,
&c., &c., all of which I have successively and successfully dealt with.
It is not my intention in this paper to describe the processes
adopted for the elimination of these several metals, but to confine
* See Table of Analyses lierewitli.
92 Mr. E. Matthey. [Feb. 10,
myself to tlie separation of lead, the presence of wMeh especially
presented at first great difficulties.
As stated above, I have found that I can separate one by one the
metals mentioned above, all of which have been associated with crude
bismuth which has come under my notice. In this, success though
gradual, has been complete : but I was still confronted by the fact
that the lead alloy was retained by the bismuth with a most charac-
teristic persistency which seemed to defy all efforts of separation
excepting by tedious wet or acid processes.
The amount of lead existing in the bismuth I operated upon, after
freeing it by dry processes from its other impurities, varied from 2 to
10 per cent.
Bearing in mind the respective fusing points of lead and bismuth,
it occurred to me that, as alloys of bismuth and lead fuse at a
temperature considerably lower than that of bismuth itself, separation
would possibly take place between the two metals at a certain point
of cooling ; I therefore made the following experiment : —
Taking a quantity of bismuth (about 10 cwt.), holding 11*5 per
cent, of lead, and fusing same, I allowed the metal to cool until the
major part of it had crystallised, then removing the fluid portion.
The residue showed by assay only 6*35 per cent, of lead, pointing
at once to the partial separation I had hoped for.
These crystals again similarly treated showed only 3*75 per cent,
The operation repeated gave crystals with only 2 per cent, of lead,
and a fourth crystallisation brought this down to below 0*5 per cent.
As a matter of possible interest, I subjoin the progressive results
during the crystallising operations of several lots up to the point of
bulking, and of finally separating every trace of lead : —
Bismuth holding 14*6 per cent. Lead.
1st crystallisation gave crystals holding 9*8 per cent, of lead.
Bismuth holding 12 per cent. Lead.
1st crystallisation gave crystals holding 6*2 per cent, of lead.
2nd „ „ „ 4*2
1887.] Contributions to the Metallurgy of Bismuth, 93
Bismuth holding 7*6 per cent. Lead,
1st crystallisation gave crystals holding 4*8 per cent, of lead.
2na ,, ,, ,, o*o ,, ,,
4tli „ „ „ 0-4 „
Bismuth holding 11 per cent. Bead.
1st crystallisation gave crystals holding 5*5 per cent, of lead.
Jna 5, ,, ,5 Z'o ,, ,,
•orcl 5, ,, ,, X*U
Bismuth holding 5*6 per cewl Bead.
1st crystallisation gave crystals holding 2*0 per cent, of lead,
2nd „ „ „ 07
3rd „ „ under 0*5
Bismuth holding 5*3 per cen^. Bead.
1st crystallisation gave crystals holding 1*8 per cent, of lead.
2nd 55 „ ,, 0*6 ,, ,,
3rd 55 55 under 0*5 „ ,,
Having attained this point, I worked upon several large quantities
of metal^ — ^with practically the same results— finally succeeding by a
continuation of the process in eliminating every trace of lead.
By the above it will be seen that the process becomes an ex-
ceedingly simple one, large quantities being treated at one time,
involving little or no loss, and occupying bours, instead of possibly
To illustrate the facilities of the separation of lead and bismuth,
alloys, I give the following figures from metal holding originally 1a.YQ
per cent, of lead.
10,675 lbs. produced, in the course of six to seven crystallisations,
9306 lbs. of available bismuth, the residue 1188 lbs. holding 40 per
cent, of lead, so that from a quantity of nearly 5 tons of bismutb and
lead alloy only about half a ton remained, holding practically the
whole of tbe lead ; the bulk of the bismuth separated by simple crys-
tallisation holding traces only of lead, which, if necessary, could be
readily eliminated by further crystallisation. From these facts it is
apparent that the separation of these two metals can be effected by
turning to account their relative fusing ^points.
94 Mi% E. B. Potilton. On a special Colour-lielation [Feb. 10,
Recapitulation of foregoing Experiment.
10,675 lbs. leady bismutli, liolding five per cent, lead, yielded
9306 lbs. of good commercial bismuth bj the crystallization process,
or within six per cent, of the total contents of pure bismuth.
Leaving for subsequent treatment —
Of alloy, holding 40 per cent, of lead, 1188 lbs., which is equal to
11*13 per cent, of the whole weight of metal treated.
Average Analysis of the Bismuth
Bismuth . . . . ,
• ♦ • ♦
• • « • •
« • t »
» • * • •
Antimony . . .
Tellurium . . .
Manganese . . .
.A-iumma . * . * .
Magnesia . . .
Insoluble earthy matter, cbiefly silica
Oxygen in combination and loss. . . . .
• • • • • e «
* # * • •
« « • • «
« « » «r •
• » • 9 e * *
« « • • •
res vjorhecl upon,
• •*»••« \J O^t*
« • « «
II. ** An Inquiry into the Cause and Extent of a special Colonr-
Eelation between certain exposed Lepidopterons Pupge
and the Surfaces which immediately surround them." By
Edward B. Poulton, M.A., of Jesus and Keble Colleges,
Oxford, Lecturer in Zoology and Comparative Anatomy
at St. Mary's Hospital, Paddington. Communicated by
Professor E. Kay Lankester, F.E.S. Received Feb-
ruary 10, 3887.
HistoriGa,l,-~-'Mv, T. W. Wood first called attention to the colour-
relation in pupae (^Entom. Soc. Proc.,' 1867, p. xcix), adducing